Oil sands are a type of unconventional petroleum deposit. The sands contain naturally occurring mixtures of sand, clay, water, and a dense and extremely viscous form of petroleum technically referred to as “bitumen,” but which may also be called heavy oil or tar. Many countries in the world have large deposits of oil sands, including the United States, Russia, and the Middle East, but the world's largest deposits occur in Canada and Venezuela.
Bitumen is so heavy and viscous (thick) that it will not flow unless heated or diluted with lighter hydrocarbons. At room temperature, bitumen is much like cold molasses, and the viscosity can be in excess of 1,000,000 cP.
Due to their high viscosity, these heavy oils are hard to mobilize, and they generally must be made to flow in order to produce and transport them. One common way to heat bitumen is by injecting steam into the reservoir. Steam Assisted Gravity Drainage (SAGD) is the most extensively used technique for in situ recovery of bitumen resources in the McMurray Formation in the Alberta Oil Sands.
In a typical SAGD process, two horizontal wells are vertically spaced by 4 to 10 meters (m). The production well is located near the bottom of the pay and the steam injection well is located directly above and parallel to the production well. In SAGD, steam is injected continuously into the injection well, where it rises in the reservoir and forms a steam chamber.
With continuous steam injection, the steam chamber will continue to grow upward and laterally into the surrounding formation. At the interface between the steam chamber and cold oil, steam condenses and heat is transferred to the surrounding oil. This heated oil becomes mobile and drains, together with the condensed water from the steam, into the production well due to gravity segregation within steam chamber.
This use of gravity gives SAGD an advantage over conventional steam injection methods. SAGD employs gravity as the driving force and the heated oil remains warm and movable when flowing toward the production well. In contrast, conventional steam injection displaces oil to a cold area, where its viscosity increases and the oil mobility is again reduced.
Although quite successful, SAGD does require enormous amounts of water in order to generate a barrel of oil. Some estimates provide that 1 barrel of oil from the Athabasca oil sands requires on average 2 to 3 barrels of water, although with recycling the total amount can be reduced to 0.5 barrel. In addition to using a precious resource, additional costs are added to convert those barrels of water to high quality steam for down-hole injection. Therefore, any technology that can reduce water or steam consumption has the potential to have significant positive environmental and cost impacts.
Additionally, SAGD is less useful in thin stacked pay-zones, because thin layers of impermeable rock in the reservoir can block the expansion of the steam chamber leaving only thin zones accessible and leaving much of the oil in other layers in place.
Indeed, in a paper by Shin & Polikar (2005), the authors simulated reservoir conditions to determine which reservoirs could be economically exploited. The simulation results showed that for Cold Lake-type reservoirs, a net pay thickness of at least 20 meters was required for an economic SAGD implementation. A net pay thickness of 15 m was still economic for the shallow Athabasca-type reservoirs because of the high permeability of this type of reservoir, despite the very high bitumen viscosity at reservoir conditions. In Peace River-type reservoirs, net pay thicker than 30 meters was expected to be required for a successful SAGD performance due to the low permeability of this type of reservoir. The results of the study indicate that the shallow Athabasca-type reservoir, which is thick with high permeability (high k×h), is a good candidate for SAGD application, whereas Cold Lake and Peace River-type reservoirs, which are thin with low permeability, are not as good candidates for conventional SAGD implementation.
Other steam based techniques include cyclic steam stimulation (CSS) and steam drive (SD), and these can be more suitable for thin or stacked pay-zones separated by impermeable layers since they often use vertical wells, providing fluid connection through heavily stratified reservoirs.
In a SD, sometimes known as a steam flood, some wells are used as steam injection wells and other wells are used for oil production. The wells can be either vertical or horizontal, but most steam floods are illustrated using vertical wells.
Two mechanisms are at work to improve the amount of oil recovered. The first is to heat the oil to higher temperatures and to thereby decrease its viscosity so that it more easily flows through the formation toward the producing wells. A second mechanism is the physical displacement that occurs in a manner similar to water flooding, in which oil is meant to be pushed to the production wells by the oncoming steam. While more steam is needed for this method than for the cyclic method, it is typically more effective at recovering a larger portion of the oil.
CSS, also known as the “Huff-and-Puff” method, consists of 3 stages: injection, soaking, and production. Steam is first injected into a well for a certain amount of time to heat the oil in the surrounding reservoir to a temperature at which it flows. After it is decided enough steam has been injected, the steam is usually left to “soak” for some time after (typically not more than a few days). Then oil is produced out of the same well, at first by natural flow (since the steam injection will have increased the reservoir pressure) and then by artificial lift. Production will decrease as the oil cools down, and once production reaches an economically determined level the steps are repeated again.
The process can be quite effective, especially in the first few cycles. However, it is typically only able to recover approximately 20% of the Original Oil in Place (OOIP), compared to steam assisted gravity drainage, which has been reported to recover over 50% of OOIP. It is quite common for wells to be produced in the cyclic steam manner for a few cycles before being put on a steam drive regime.
One concept for improving production is the “multilateral” or “fishbone” well configuration idea. The concept of fishbone wells for non-thermal horizontal wells was developed by Petrozuata in Venezuela starting in 1999. That operation was a cold, viscous oil development in the Faja del Orinoco Heavy Oil Belt. The basic concept was to drill open-hole side lateral wells or “ribs” off the main spine of a producing well prior to running slotted liner into the spine of the well. Such ribs appeared to significantly contribute to the productivity by increasing the area of reservoir contact of the wells when compared to wells without the ribs in similar geology. A variety of multilateral well configurations are possible, although many have not yet been tested.
The advantages of multilateral wells can include:
1) Higher Production. In the cases where thin pools are targeted, vertical wells yield small contact with the reservoir, which causes lower production. Drilling several laterals in thin reservoirs and increasing contact improves recovery. Slanted laterals can be of particular benefit in thin stacked pay zones.
2) Decreased Water/Gas Coning. By increasing the length of “wellbore” in a horizontal strata, the inflow flux around the wellbore can be reduced. This allows a higher withdrawal rate with less pressure gradient around the producer. Coning (literally a cone of water in the region of the producer) is aggravated by pressure gradients that exceed the gravity forces that stabilize fluid contacts (oil/water or gas/water), so that coning is minimized with the use of multilaterals because they minimize the pressure gradient.
3) Improved sweep efficiency. By using multilateral wells, the sweep efficiency may be improved, and/or the recovery may be increased due to the additional area covered by the laterals mitigating the natural heterogeneity in the reservoir.
4) Faster Recovery. Production from the multilateral wells is at a higher rate than that in single vertical or horizontal wells, because the reservoir contact is higher in multilateral wells.
5) Decreased environmental impact. The volume of consumed drilling fluids and the generated cuttings during drilling multilateral wells are less than the consumed drilling fluid and generated cuttings from separated wells, at least to the extent that two conventional horizontal wells are replaced by one dual lateral well and to the extent that laterals share the same mother-bore. The surface footprint is also smaller, as only one location is required. Therefore, the impact of the multilateral wells on the environment can be reduced.
6) Saving time and cost. Drilling several laterals in a single well may result in time and cost saving in comparison with drilling several separate wells in the reservoir.
Although an improvement, the multilateral well methods have disadvantages too. One disadvantage is that fishbone wells are more complex to drill and clean up. Indeed, some estimate that multilaterals cost about 20% more to drill and complete than conventional slotted liner wells. Another disadvantage is increased risk of accident or damage, due to the complexity of the operations and tools.
Sand control can also be difficult. In drilling multilateral wells, the mother well bore can be cased to control sand production, however, the legs branched from the mother well bore are usually open hole. Therefore, the sand control from the branches is not easy to perform. There is also increased difficulty in modeling and prediction due to the sophisticated architecture of multilateral wells.
Another area of uncertainty with the fishbone concept is whether the ribs will establish and maintain communication with the steam chambers, or will the open-hole ribs collapse and block flow. One of the characteristics of the Athabasca Oil Sands is that they are unconsolidated sands that are bound by the million-plus centipoises bitumen. When heated to 50-80° C. the bitumen becomes slightly mobile. At this point the open-hole rib could collapse. If so, flow would slow to a trickle, temperature would drop, and the rib would be plugged. However, if the conduit remains open at least long enough that the bitumen in the near vicinity is swept away with the warm steam condensate before the sand grains collapse, then it may be possible that a very high permeability, high water saturation channel might remain even with the collapse of the rib. In this case, the desired conduit would still remain effective.
Another uncertainty with many ribs along a fishbone producer of this type is that one rib may tend to develop preferentially at the expense of all the other ribs leading to very poor conformance and poor overall results. This would imply that some form of inflow control may be warranted to encourage more uniform development of all the ribs.
Multilateral wells have been used for a variety of patented methods. EP2193251 discloses a method of drilling multiple short laterals that are of smaller diameter. These multiple short laterals can be drilled at the same depth from the same main wellbore, so as to perform treatments in and from the small laterals to adapt or correct the performance of the main well, the formation properties, the formation fluids and the change of porosity and permeability of the formation. However, this method does not increase overall reservoir contact, nor improve injectivity, nor increase well-to-well fluid communication.
US20110036576 discloses a method of injecting a treatment fluid through a lateral injection well such that the hydrocarbon can be treated by the treatment fluid before production. However, the addition of treatment fluid is known in the field and this well configuration does not increase the contact with the hydrocarbon reservoir.
CA2684049 describes the use of infill wells (between pairs of SAGD well-pairs) that are equipped with multilateral wells, so as to allow the targeting of additional regions. However, no general applicability to SAGD was described in this application.
Pham and Stalder further developed the fishbone well idea to allow increased application for SAGD processes. U.S. Ser. Nos. 61/825,945, filed May 21, 2013, and 61/826,329, filed May 22, for example, describe general application of fishbone wells in SAGD, as well as developing a radial fishbone SAGD well configuration. Both disclosures allow increased contact with the reservoir, increased injectivity and further, the unique patterns reduced overall well numbers and well-pad costs. However, the well configurations shown therein are optimized for use with horizontal wells and gravity drainage, and not for the steam drive mechanisms of CSS and SD.
Therefore, although beneficial, the multilateral well concept could be further developed to address some of these disadvantages or uncertainties. In particular, a method that combines multilateral well architecture with steam drive processes and/or huff-and-puff processes would be beneficial, especially if such methods conserved the water, energy, and/or cost to produce a barrel of oil.